US5163162A - System and method for data recovery in multiple head assembly storage devices - Google Patents

System and method for data recovery in multiple head assembly storage devices Download PDF

Info

Publication number
US5163162A
US5163162A US07/612,661 US61266190A US5163162A US 5163162 A US5163162 A US 5163162A US 61266190 A US61266190 A US 61266190A US 5163162 A US5163162 A US 5163162A
Authority
US
United States
Prior art keywords
head
data
misaligned
storage device
read
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
US07/612,661
Other languages
English (en)
Inventor
Robert L. Berry
Brandt C. Centerwall
Stephen G. Luning
Forrest L. Wade
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
International Business Machines Corp
Original Assignee
International Business Machines Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by International Business Machines Corp filed Critical International Business Machines Corp
Priority to US07/612,661 priority Critical patent/US5163162A/en
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, A CORP. OF NY reassignment INTERNATIONAL BUSINESS MACHINES CORPORATION, A CORP. OF NY ASSIGNMENT OF ASSIGNORS INTEREST. Assignors: WADE, FORREST L., LUNING, STEPHEN G., BERRY, ROBERT L., CENTERWALL, BRANDT C.
Priority to CA002050161A priority patent/CA2050161C/en
Priority to MYPI91001891A priority patent/MY108075A/en
Priority to CN91109643A priority patent/CN1023518C/zh
Priority to KR1019910018004A priority patent/KR950013379B1/ko
Priority to JP3303861A priority patent/JPH0792896B2/ja
Priority to EP19910118321 priority patent/EP0485795A3/en
Publication of US5163162A publication Critical patent/US5163162A/en
Application granted granted Critical
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B19/00—Driving, starting, stopping record carriers not specifically of filamentary or web form, or of supports therefor; Control thereof; Control of operating function ; Driving both disc and head
    • G11B19/02—Control of operating function, e.g. switching from recording to reproducing
    • G11B19/04—Arrangements for preventing, inhibiting, or warning against double recording on the same blank or against other recording or reproducing malfunctions
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B20/00—Signal processing not specific to the method of recording or reproducing; Circuits therefor
    • G11B20/10—Digital recording or reproducing
    • G11B20/18—Error detection or correction; Testing, e.g. of drop-outs
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B21/00—Head arrangements not specific to the method of recording or reproducing
    • G11B21/02—Driving or moving of heads
    • G11B21/08—Track changing or selecting during transducing operation
    • G11B21/081—Access to indexed tracks or parts of continuous track
    • G11B21/083—Access to indexed tracks or parts of continuous track on discs
    • G11B21/085—Access to indexed tracks or parts of continuous track on discs with track following of accessed part
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B20/00—Signal processing not specific to the method of recording or reproducing; Circuits therefor
    • G11B20/10—Digital recording or reproducing
    • G11B20/18—Error detection or correction; Testing, e.g. of drop-outs
    • G11B20/1816—Testing
    • G11B2020/183—Testing wherein at least one additional attempt is made to read or write the data when a first attempt is unsuccessful
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B5/00—Recording by magnetisation or demagnetisation of a record carrier; Reproducing by magnetic means; Record carriers therefor
    • G11B5/48—Disposition or mounting of heads or head supports relative to record carriers ; arrangements of heads, e.g. for scanning the record carrier to increase the relative speed
    • G11B5/488—Disposition of heads
    • G11B5/4886—Disposition of heads relative to rotating disc
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B5/00—Recording by magnetisation or demagnetisation of a record carrier; Reproducing by magnetic means; Record carriers therefor
    • G11B5/48—Disposition or mounting of heads or head supports relative to record carriers ; arrangements of heads, e.g. for scanning the record carrier to increase the relative speed
    • G11B5/54—Disposition or mounting of heads or head supports relative to record carriers ; arrangements of heads, e.g. for scanning the record carrier to increase the relative speed with provision for moving the head into or out of its operative position or across tracks
    • G11B5/55—Track change, selection or acquisition by displacement of the head
    • G11B5/5521—Track change, selection or acquisition by displacement of the head across disk tracks
    • G11B5/5526—Control therefor; circuits, track configurations or relative disposition of servo-information transducers and servo-information tracks for control thereof
    • G11B5/553—Details
    • G11B5/5547—"Seek" control and circuits therefor

Definitions

  • the present invention is a system and method for detecting and correcting errors in a multiple head storage device. More specifically, this invention relates to a system and method for detecting errors caused by the misalignment of one or more heads in a multiple head storage device, and for manipulating these heads in order to recover undecipherable data caused by the misalignment.
  • mass-storage devices have traditionally lagged significantly behind the speed of state-of-the-art computers. This is largely due to the fact that, unlike the computational portion of a computer, mass-storage devices contain moving parts. Consequently, there is a continuing need for faster data storage devices which can effectively work in conjunction with today's high speed computers.
  • computer storage devices for storing data used in conjunction with the computational portion (also referred to as the "host" portion) of a computer typically have one or more read/write or read-only heads which pass over some medium upon which data is stored.
  • the storage capabilities of these mediums are usually based upon magnetic or optical principles.
  • the faster devices using such mediums are typically in the form of magnetic or optical disks.
  • the heads and their associated controlling mechanisms are configured so that each head will only read from and write to certain portions of a storage medium. These portions are known as tracks. On a magnetic or optical disk storage device, these tracks are concentric, having some amount of concentric non-used space between each track. When a host device or host computer requests that certain data residing on one or more tracks be sent to it, the head is then manipulated over those desired tracks in order to read the desired data.
  • the number and spacing of these tracks used with the above-noted storage mediums are generally dictated by some combination of hardware and software which directly or indirectly controls the heads.
  • the width of the track itself is a function of the head.
  • the result is the development of storage devices in which data is packed more densely, and in which the medium is passed under the heads more quickly.
  • data when data is sent to one of these multiple head storage devices from a host computer, it is distributed or "striped" into sections evenly amongst the heads of the storage device. For example, in a three-head storage device, the data might be split up so that first two bytes might go to head 1, the next two to head 2, the next two to head 3, and then the next two to head 1 again. In this example, each section is two bytes. During this process, each head is positioned over a different track (each track may or may not reside on a different storage medium, as will be explained regarding FIGS. 1 and 2 below). This striping operation continues until all of the data sent by the host computer has been written to the storage device.
  • the multiple head storage device uses only one head.
  • a problem that occurs in storage devices generally is that the head sometimes becomes misaligned with a track. This misalignment causes the head to read useless or "undecipherable" data.
  • One method for realigning the head is to move the head back and forth by small increments until the head is again aligned with the track so that the data from that track becomes decipherable, and can be recovered.
  • the head is moved in increments sufficiently small and in such a pattern so that it can seek out the target track without wandering too far from the target track (or too close to another track).
  • the device will continue to be used, or else the data will be recovered, and the device replaced.
  • the problem of misalignment is amplified by the fact that there is more than one head to consider.
  • the heads of multiple head storage devices are typically locked together as a multiple head assembly so that they all move in unison. In this way, misalignments usually affect all of the heads, and thus corrective measures can be taken as though there was only a single head to consider.
  • FIG. 1 is used to illustrate the problems with error detection and data recovery that can occur in a multiple head storage device described above.
  • this figure shows an example of a multiple head storage device 104 having three separate disks, with one head servicing each disk.
  • the heads are attached together to form a multiple head assembly 106 which constrains all of the heads to move in unison.
  • heads in the example shown in FIG. 1 may be initially set at the proper distances for them to read or write properly on their designated tracks, it is still possible for one or more of the heads to become misaligned.
  • head 3 is shown to be misaligned.
  • ECC Error Correction Code
  • the ECC is conventionally used to check some amount of pre-striped data (that is, it is used to check the data after it is interleaved), it would be difficult to determine from an error detected from the occurrence of undecipherable data specifically which head caused the error.
  • head 3 in multiple head storage device 104 Another problem is that by moving head 3 in multiple head storage device 104 by a small increment to correct the misalignment of head 3, this will typically cause heads 1 and 2 to become misaligned. This would cause any data read from heads 1 and 2 to become undecipherable. Thus, the data from these other heads must be taken care of either before or after the multiple assembly 106 is moved by small increments, since data from all heads need to be interleaved before the data can be sent to a host computer 102. Where at least one head is misaligned, the fact that all the heads cannot simultaneously read decipherable data makes this interleave process more difficult.
  • FIG. 2 shows another example of a multiple head storage device. This type of multiple head storage device utilizes more than one head per disk.
  • the example shown in FIG. 2 shows such a multiple head storage device 104 having three heads and n tracks, where n would typically be an integer divisible by 3.
  • the present invention overcomes the deficiencies of the related art noted above by providing a system and method for detecting and correcting misaligned heads in a multiple head storage device.
  • an embodiment of the present invention first analyzes a block of interleaved data received from a multiple head storage device, and determines if there are any errors (that is, undecipherable data) indicating a misalignment of one or more of the heads of the multiple head storage device. In addition, if an error is detected, the present invention determines which head is misaligned.
  • the block of interleaved data is sent to a data buffer, and error detection signals are sent to status bits.
  • the status bits are used as flags to indicate whether any misaligned heads exist, and which head(s) is misaligned.
  • Data recovery control logic analyzes the status bits to check if any misaligned heads exist. If there are none, then the block of interleaved data in the data buffer can be sent to a host computer, rewritten to the multiple head storage device, or sent to some other device.
  • the data recovery control logic sends control signals to the multiple head storage device indicating that the multiple head storage device should move the misaligned head by small increments and send data from the misaligned head to the present invention. This continues to occur until the status bits indicate that no error is detected (that is, the misaligned head has become aligned, and the data from the head has become decipherable).
  • the data read from the aligned head is interleaved with the block of interleaved data in the data buffer. This is done so that the portion of the interleaved block of data which had been read from the misaligned head is replaced with data read from the aligned head. This data can then be sent to any of the devices noted above.
  • the present invention can detect the existence of a misaligned head from a block of interleaved data read from a multiple head storage device, and also re-align the head and recover decipherable data before the block of interleaved data is sent to the host computer.
  • FIG. 1 is a block diagram of a multiple head storage device having 3 disks and one head for each disk, where head 3 is shown to be misaligned.
  • FIG. 2 is a top view of a multiple head storage device having a plurality of tracks and three heads, where head 3 is shown to be misaligned.
  • FIG. 3 is a high-level block diagram of an embodiment of the present invention.
  • FIG. 4 is a block diagram of an embodiment of the present invention with a temporary data buffer.
  • FIG. 5 is a block diagram of an embodiment of the present invention having separate error detection devices and an error control mechanism.
  • FIG. 6 is a block diagram of an embodiment of the present invention having separate error detection devices and showing the flow of data thereto.
  • FIG. 7 is a block diagram of an alternate embodiment of the temporary data buffer.
  • FIG. 8 is a flow diagram of the operation of an embodiment of the present invention.
  • FIG. 9 is a flow diagram of the operation an embodiment of the present invention having re-read capabilities.
  • the present invention is a system and method for detecting and correcting errors in a multiple head storage device. More specifically, the present invention relates to a system and method for error detection and data recovery in a multiple head storage device where one or more of the heads can become misaligned. This is done by detecting which head(s) is misaligned with its target track(s), buffering the data read from the head(s) which was not misaligned, incrementally moving the misaligned head(s) until it becomes aligned with its target track(s) so that it can read decipherable data, and interleaving the decipherable data from the aligned head(s) into the buffered data so as to produce a block of data for, for example, the host computer.
  • the present invention detects errors (that is, undecipherable data) in blocks of interleaved data caused by the misalignment of one or more heads in a multiple head storage device, and recovers the data from the target tracks of these heads.
  • the recovered data is typically sent to the host computer. Once the data is recovered, the multiple head storage device can then be replaced, and the data can be written to a new multiple head storage device.
  • the data received by the host computer can be rewritten to the original multiple head storage device, since the data will be written to the locations of the storage medium corresponding to the positions of the heads.
  • the data when data is rewritten to a misaligned head whose position has moved from its original position by some amount, the position of the rewritten target track will have moved from the position of the original track by a corresponding amount.
  • the data can be read in a normal fashion.
  • the present invention contemplates that the block of interleaved data can be rewritten to the original multiple head storage device without first sending the data to the host computer.
  • FIG. 3 A broad concept of the present invention can best be described with regard to FIG. 3.
  • the embodiment shown in FIG. 3, as well as others contemplated by the present invention utilizes a multiple head storage device 104 having two or more heads for reading data or writing data simultaneously. It is further contemplated that the multiple head storage device 104 automatically splits up (stripes) data into sections when data is sent to it from a host device such as a host computer 102. Conversely, when data is read from the multiple head storage device 104, it is contemplated that the multiple head storage device 104 interleaves the data back into the same form in which it was originally sent by the host computer 102.
  • multiple head storage devices 104 Examples of the type of multiple head storage devices 104 contemplated above are those shown in FIGS. 1 and 2.
  • a more specific example of such a multiple head storage device 104 is a 3370 multiple disk DASD from IBM Corporation of Armonk, N.Y. It should be understood, however, that the present invention also contemplates operation with multiple head storage devices 104 which do not stripe and interleave data automatically.
  • the present invention accepts a block of interleaved data sent along data line 306 to an error detection device 302 for detecting errors in the data caused by a misalignment of one or more heads. Regardless of whether any errors are detected by error detection device 302, this block of interleaved data is automatically sent via data line 310 to data buffer 304.
  • this block of data is, for example, 4 Kilobytes in size, although any convenient size can be used.
  • Data buffer 304 is broken up in FIG. 3 for illustrative purposes into sections to indicate which of the heads from multiple head storage device 104 read that particular section of data. For example, those sections of data buffer 304 containing a 1 represent data which had been read from head 1. Similarly, a 2 represents the data read from head 2. When the n th section of data has been read from head n, the sequence begins anew. In effect, data buffer 304 shows the source of each section of the block of interleaved data with regard to which head read the data.
  • error detection signals 308 are sent from error detection device 302 to status bits 318. These status bits 318 are used as a flagging mechanism to indicate which heads of the multiple head storage device 104 are misaligned. If the status bits 318 indicate that no heads are misaligned, then the data in data buffer 304 can be sent to the host computer 102 via data line 316.
  • the block of interleaved data received from multiple head storage device 104 is still sent to the data buffer 304.
  • the status bits 318 will indicate which heads are misaligned. For example, if an error has been detected in head 2 of multiple head storage device 104, then the error detection device 302 will send a signal to status bit S2 of status bits 318, and set that bit to indicate that head 2 is misaligned.
  • the block of interleaved data stored in data buffer 304 will not be sent to host computer 102. This is because the data which was read from head 2 is undecipherable, and decipherable data from head 2 must first be recovered. Since the data read into the data buffer 304 from the other heads is decipherable, the present invention will preserve this decipherable data in the data buffer 304 so that it can interleave the recovered data from head 2 with the decipherable data.
  • data recovery control logic 320 is responsive to the status bits 318 when the status bits 318 indicate that a head is misaligned. This indication is sent via control line 312. Thus, the fact that status bit S2 is set informs the data recovery control logic 320 that head 2 is misaligned.
  • the error recovery control logic 320 Upon receipt of an indication from the status bits 318 that head 2 is misaligned, the error recovery control logic 320 sends control signals to the multiple head storage device 104 via control line 314, causing multiple head assembly 106 (see FIGS. 1 and 2) to move traversely to the length of the target track by some small increment In one embodiment, all of the heads of the multiple head storage device 104 read data, and the multiple head storage device 104 sends a new block of interleaved data to the error detection device 302 via data line 306.
  • Status bit S2 is then checked by the data recovery control logic 320 to determine if the data read from head 2 of multiple head storage device 104 is still undecipherable. If the data is still undecipherable, then the data recovery control logic 320 will again cause multiple head assembly 106 to move by a small increment, and another block of interleaved data will be sent to the error detection device 302.
  • This data recovery cycle will repeat itself until status bit S2 of status bits 318 indicates that the data from head 2 is decipherable, thus indicating the alignment of head 2.
  • the decipherable data from head 2 is interleaved into its proper locations in the block of interleaved data within the data buffer 316.
  • those sections of the data buffer 316 marked with a 2 will be over-written by the decipherable data read from the now-aligned head 2.
  • the present invention contemplates that this interleaving can be controlled by the data recovery control logic 320, or it can be controlled by some separate interleaving means.
  • the data in data buffer 304 can be sent as a block to host computer 102. If more than one head is misaligned, then the data recovery cycle is repeated for as many heads as are out of alignment before the data in data buffer 304 is sent to host computer 102.
  • the present invention contemplates usage with multiple head storage devices 104 in which peripheral devices are necessary for the striping and interleaving operations.
  • the present invention can also be integrated into a multiple head storage device 104 having improved error detection and data recovery.
  • the present invention contemplates usage with various types of read/write and read-only disk storage devices, including magnetic and optical disks.
  • the present invention contemplates using a data buffer 304 and corresponding status bits 318 which are large enough to hold multiple blocks of interleaved data, and thus allow more than one block of interleaved data to be read from the multiple head storage device 104 at one time. In this way, more than one consecutive block of interleaved data can be checked for errors and sent to the host computer 102 at a time.
  • FIG. 4 shows a specific implementation of the present invention using a dual-buffer approach in order to accomplish the above-noted objectives of recovering data from a misaligned head in a multiple head storage device.
  • FIG. 4 is best explained by example, again using the example where head 2 of multiple head storage device 104 is misaligned.
  • the X's in sections of data buffer 304 indicate that the data read into data buffer 304 from head 2 is undecipherable.
  • status bit S2 of status bits 318 is set to indicate that head 2 is misaligned.
  • the set status bit S2 is detected by data recovery control logic 320, which then causes at least head 2 to move by small increments in an effort to re-align head 2 over its target track(s), and recover the data from that track(s).
  • One solution of the present invention for overcoming this problem is to permit only data read by the misaligned head (in this case, head 2) to be read into the data buffer 304.
  • Implementation of this approach entails extracting only those sections from the interleaved data of data line 310 read from head 2. Then, the data from head 2 is interleaved into its appropriate locations (in this example, the locations marked with an X) within the data buffer 304. The extraction and interleaving would need to be implemented for each read from the multiple head storage device 104 during the recovery process, and thus would continue until status bits 318 indicated that head 2 was aligned.
  • FIG. 4 Another embodiment as shown in FIG. 4 includes the use of temporary status bits 406, and a temporary data buffer 402 into which a block of interleaved data is read when status bits 318 indicate that one or more heads of the multiple head storage device 104 are misaligned.
  • blocks of interleaved data are read into the temporary data buffer 402 only after the data recovery cycle outlined above has begun.
  • error detection signals 308 are sent to the temporary status bits 406.
  • the data recovery control logic 320 checks the temporary status bits 406 rather than the status bits 318 to determine whether a misaligned head has become an aligned head.
  • the data recovery control logic 320 sends control signals via control line 404 to interleave the data read into the temporary data buffer 402 from aligned head 2 to the appropriate sections (in FIG. 4, marked by an X) of the data buffer 304. This causes the undecipherable data in the data buffer 304, which was previously read by head 2, to consequently be overwritten. The other data that was read into the temporary data buffer 402 during the data recovery cycle is ignored.
  • the entire block of interleaved data in data buffer 304 is decipherable. This block of interleaved data can the be sent to the host computer 102.
  • An advantage of using the temporary data buffer 402 as described above is that it allows data to be read directly into a the temporary data buffer 402 during the data recovery process. This is in contrast to having to carefully separate the data read from the misaligned head from the rest of the block of interleaved data before interleaving it into data buffer 304 each time a block of interleaved data is sent from the multiple head storage device 104 during a data recovery cycle.
  • the use of the temporary data buffer allows the present invention to simply replace the undecipherable data in the data buffer 304 with decipherable data in the temporary data buffer 402, rather than request that the multiple storage device 104 again read data once the temporary status bits 406 indicate that the misaligned head has become aligned.
  • the multiple head storage device 104 would have to again read the data from the now-aligned head.
  • temporary data buffer 402 indicates that data from all the heads (that is, the entire block of interleaved data) is read into the temporary data buffer 402, even though only that data from the misaligned head is scrutinized during the data recovery cycle. However, it should be understood that the present invention also contemplates an embodiment where temporary data buffer 402 permits only data from a misaligned head to be read in.
  • FIG. 4 allow for an efficient way to recover data and improve the integrity of the block of interleaved data in the data buffer 304 before it is sent to a device such as the host computer 102.
  • FIG. 5 shows another embodiment of the present invention.
  • the status bits 318 and temporary status bits 406 are shown to be part of data buffer 304 and temporary data buffer 402, respectively. This allows the status bits 318 and temporary status bits 406 to be set by status data generated by the error detection device(s) 302, and sent to the status bits 318 and temporary status bits 406 via data line 310.
  • This status data is sent either to the status bits 318 or to temporary status bits 406 after a block of interleaved data is sent to either data buffer 304 or temporary data buffer 402, respectively.
  • the data buffer 304 and temporary data buffer 402 are used to store the data from the multiple head storage device, as well as to store the status bits 318 and temporary status bits 406, respectively.
  • FIG. 5 also shows that the present invention contemplates the use of more than one error detection device 302.
  • An error detection device 302 can be used for each head that exists in the multiple head storage device 104, or one error detection device 302 can be used for any number of heads. In any event, each error detection device 302 generates an error detection signal 308, which is detected by the appropriate status bits.
  • An advantage to using more than one error detection device 302 is that it allows the burden of error detection to be split up among several different components. This is important when the multiple head storage device 104 is capable of sending data at very high speeds. Thus, using more than one error detection device 302 allows each of the error detection devices 302 to each operate at a fraction of the speed that a single error detection device 302 would be required to operate at if used by itself.
  • each error detection device 302 would need to operate at where more than one error detection device 302 is used would depend upon the number of heads and the number of error detection devices 302 used. For example, if there are three heads in a multiple head storage device 104 and three error detection devices 302, each of the three error detection devices 302 would need to operate at one-third of the speed that a single error detection device 302 would operate at if used by itself. This allows for an easier and less expensive design of the error detection devices 302.
  • an ECC needs to have been previously generated for each section of data that has been striped by, and written to, the multiple head storage device 104.
  • each striped section of data written by each head of the multiple head storage device 104 needs to have its own ECC.
  • One way of doing this as contemplated by the present invention is by striping the data prior to sending it to the multiple head storage device 104 into the same number and length of sections as the multiple head storage device 104 will do when the multiple head storage device 104 receives the data. Then, generate an ECC for each section of the data, interleave the data, and send it to the multiple head storage device 104 to be stored.
  • the same logic for example, the same ECC algorithm
  • the error detection devices 302 of the present invention can be used by the error detection devices 302 of the present invention in reverse to check the ECC and determine if any errors occurred.
  • This two-way use of the same ECC logic is shown by FIG. 6.
  • the logic chips used for generating and checking ECCs generally are readily available, and one typical example is from Simulex Corporation of Tustin, Calif.
  • FIG. 7 shows an alternative embodiment of the temporary data buffer 702 and corresponding status bits 704.
  • This embodiment provides for a block of temporary data buffer space for each head used in the multiple head storage device 104. This allows for the recovery of data when more than one head is misaligned before any recovered data is interleaved back into data buffer 304.
  • the temporary data buffer of FIG. 7 is best explained by way of example. If heads 1 and 3 of the multiple head storage device 104 are both misaligned, the present invention might first recover the data from head 1 using the data recovery cycle explained above. Thus, head 1 would be moved by some small increment until head 1 block 706 contained decipherable data in the locations designated for that particular head (that is, those storage areas marked 1).
  • the present invention then begins the data recovery cycle to recover data from misaligned head 3 rather than immediately interleaving the data from head 1 block 706 with the data in the data buffer 304. Thus, the present invention will then send data to head 3 block 710 until the data for head 3 becomes decipherable. This is indicated by the temporary status bits 704 corresponding to head 3 block 710.
  • the temporary data buffer 702 in FIG. 7 shows that each block for each head has space for all the data read in from all the heads, and that each of the corresponding status bits has a bit for each head. This was done so that each relevant portion of data would not have to be extracted.
  • the data recovery control logic 320 only focusses in on that data and temporary status bits 704 that are relevant.
  • head 1 block 706 only that data read from head 1 would be relevant, as would only corresponding status bit S1.
  • head 3 block 710 only that data read from head 3 would be relevant, as would only corresponding status bit S3. This approach helps improve the speed of the overall invention.
  • the present invention also contemplates a temporary data buffer 702 in which only relevant data is read in.
  • the error detection devices are stored on CMOS.
  • the data buffer 304 and temporary data buffer 402 use 256 kilobytes of SRAM.
  • 2 bytes of data from each of the heads of the multiple head storage device 104 is read into the data buffer 304 at a time.
  • each of the numbers shown in FIG. 5 representing a section of data written by one of the heads represents 2 bytes.
  • any amount of data could be used for each section.
  • an embodiment of the present invention contemplates that there are two situations in which the existence of a "misaligned" head (that is, a head that produces undecipherable data) can be indicated.
  • the first situation is described above, in which a head is not aligned with its target track.
  • the second situation is when undecipherable data is read for reasons unrelated to the physical position of the head. These reasons include the existence of dust particles coming between the head and the target track.
  • the data recovery logic 320 in one embodiment of the present invention first sends control signals to the multiple head storage device 104 to re-read the data from the head in which an error was detected This is done without incrementally moving the head.
  • the error detection and buffering schemes indicated in the above-noted embodiments of the present invention would be used as indicated above during this re-read.
  • the data recovery control logic 320 then sends control signals to the multiple head storage device 104 to move the head by small increments, and the data recovery cycle outlined above would begin.
  • the host computer is given an indication that a critical error has occurred.
  • FIG. 8 shows the operation of the present invention.
  • a block of interleaved data is read into the data buffer 304, as indicated by a block 802.
  • the status bits 318 which indicate if an error has occurred in the data are checked, as indicated by a block 806. If no error has occurred, the block of interleaved data is sent to the host computer 102, as indicated by a decision block 808 and a block 804.
  • Another block of interleaved data is then read into the data buffer 304 as indicated by block 802.
  • the next step is to determine which head is misaligned, as indicated by a block 810. Once this has been determined, the head which is misaligned is moved by some small increment (either by itself or in conjunction with any attached or affected heads of the multiple head storage device 104) and the data from the misaligned head is read as indicated by a block 812. This is accomplished without affecting decipherable data read from the remaining heads. The status bits 318 are then checked to see if the data contains any errors, as indicated by a block 814.
  • the misaligned head is again moved by some small increment, as indicated by a decision block 816, and block 812. If the status bits 318 indicate that the misaligned head has become an aligned head, the data from the now-aligned head is then interleaved into the block of interleaved data in the data buffer 304, as indicated by decision block 816 and a block 818. The block of interleaved data can then be sent on to the host computer, as indicated by block 820.
  • FIG. 9 shows the operation of the present invention having re-read capabilities. This embodiment contemplates the operation discussed above with regard to FIG. 8, and in addition, adds steps including the re-reading of data from a misaligned head as indicated by a block 902. This re-read is accomplished without moving the misaligned head.
  • the status bits 318 are checked to see if the head is still misaligned, as indicated by a block 904. If no error has been found to occur which would indicate the existence of a misaligned head, as indicated by decision block 906, then the data is interleaved as per block 818 of FIG. 8. If, however, decision block 816 indicates that an error has occurred, then if the misaligned head has been re-read less than some set number of iterations, the misaligned head is re-read again, as indicated by a decision block 908 and block 902.
  • the misaligned head has been re-read more than the set number of iterations, then when the present invention comes to the step indicated by decision block 908, the misaligned head will be moved by small increments, as indicated by block 812 of FIG. 8.
  • the operational steps shown in FIG. 9 could be inserted into FIG. 8.
  • the present invention can be implemented in a software embodiment.
  • the various components and steps would be implemented in software to perform the functions of the present invention.
  • Any presently available or future developed computer software language can be employed in such a software embodiment of the present invention.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Signal Processing For Digital Recording And Reproducing (AREA)
  • Digital Magnetic Recording (AREA)
  • Indexing, Searching, Synchronizing, And The Amount Of Synchronization Travel Of Record Carriers (AREA)
US07/612,661 1990-11-14 1990-11-14 System and method for data recovery in multiple head assembly storage devices Expired - Fee Related US5163162A (en)

Priority Applications (7)

Application Number Priority Date Filing Date Title
US07/612,661 US5163162A (en) 1990-11-14 1990-11-14 System and method for data recovery in multiple head assembly storage devices
CA002050161A CA2050161C (en) 1990-11-14 1991-08-28 System and method for data recovery in multiple head assembly storage devices
KR1019910018004A KR950013379B1 (ko) 1990-11-14 1991-10-14 오정렬된 헤드 정렬 시스템, 오정렬된 헤드 정렬 방법, 오정렬된 헤드 검출 시스템 및 오정렬된 헤드 검출 방법
CN91109643A CN1023518C (zh) 1990-11-14 1991-10-14 多磁头组件存储器中恢复数据的系统和方法
MYPI91001891A MY108075A (en) 1990-11-14 1991-10-14 System and method for data recovery in multiple head assembly storage devices
JP3303861A JPH0792896B2 (ja) 1990-11-14 1991-10-24 誤り位置決めされたヘッドを位置決めする装置および方法
EP19910118321 EP0485795A3 (en) 1990-11-14 1991-10-28 System and method for data recovery in multiple head assembly storage devices

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US07/612,661 US5163162A (en) 1990-11-14 1990-11-14 System and method for data recovery in multiple head assembly storage devices

Publications (1)

Publication Number Publication Date
US5163162A true US5163162A (en) 1992-11-10

Family

ID=24454111

Family Applications (1)

Application Number Title Priority Date Filing Date
US07/612,661 Expired - Fee Related US5163162A (en) 1990-11-14 1990-11-14 System and method for data recovery in multiple head assembly storage devices

Country Status (7)

Country Link
US (1) US5163162A (ko)
EP (1) EP0485795A3 (ko)
JP (1) JPH0792896B2 (ko)
KR (1) KR950013379B1 (ko)
CN (1) CN1023518C (ko)
CA (1) CA2050161C (ko)
MY (1) MY108075A (ko)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5465244A (en) * 1991-03-28 1995-11-07 Kabushiki Kaisha Toshiba Disc reproducing apparatus
US5511051A (en) * 1992-08-31 1996-04-23 Olympus Optical Co., Ltd. Optical recording/reproducing apparatus capable of multi-track access for simultaneous multirack recording/reproduction
US5745639A (en) * 1995-01-04 1998-04-28 International Business Machines Corporation Apparatus for improving the continuous supply or display time of full-motion video and other data
US6314433B1 (en) * 1998-06-12 2001-11-06 Hewlett-Packard Company Frame-based heroic data recovery
US6345366B1 (en) * 1998-09-24 2002-02-05 International Business Machines Corporation Method and device for error recovery
US6381088B1 (en) * 1998-11-06 2002-04-30 Acorn Technologies, Inc. Apparatus for developing a dynamic servo signal from data in a magnetic disc drive and method
US6412088B1 (en) * 1999-12-02 2002-06-25 Maxtor Corporation Method and apparatus for using block reread
US6633521B1 (en) * 1999-08-27 2003-10-14 Fujitsu Limited Storage apparatus and servo recovery method of it
USRE40413E1 (en) * 1998-11-06 2008-07-01 Purchased Patent Management Llc Method and apparatus for developing a dynamic servo signal from data
US20100162068A1 (en) * 2008-12-19 2010-06-24 Kabushiki Kaisha Toshiba Memory device

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5353170A (en) * 1993-05-19 1994-10-04 International Business Machines Corporation Error recovery data storage system and method with two position read verification
US6706577B1 (en) * 1999-04-26 2004-03-16 Taiwan Semiconductor Manufacturing Company Formation of dual gate oxide by two-step wet oxidation
US6606041B1 (en) * 2000-05-10 2003-08-12 Micron Technology, Inc. Predictive timing calibration for memory devices

Citations (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3943566A (en) * 1974-07-17 1976-03-09 International Business Machines Corporation Dynamic skew correction for rotating head magnetic recorder
US4024498A (en) * 1975-08-04 1977-05-17 Mcintosh Billy L Apparatus for dead track recovery
US4092732A (en) * 1977-05-31 1978-05-30 International Business Machines Corporation System for recovering data stored in failed memory unit
US4103234A (en) * 1967-11-24 1978-07-25 General Dynamics Corp. System for transmission storage and/or multiplexing of information
US4148082A (en) * 1976-09-13 1979-04-03 Sony Corporation Tracking control apparatus
USRE31166E (en) * 1967-08-09 1983-03-01 Rca Corporation Recovery of normally illegible, magnetically recorded information
US4414587A (en) * 1981-12-17 1983-11-08 Sri International Skew sensing for digital tape playback
US4445145A (en) * 1980-11-07 1984-04-24 Sony Corporation Video signal reproducing apparatus
US4445153A (en) * 1980-03-07 1984-04-24 Ricoh Company, Ltd. Method of correcting track misalignment of floppy disc and apparatus therefor
US4477849A (en) * 1982-04-12 1984-10-16 Pioneer Magnetics, Inc. Magnetic disc drive head alignment system
US4516165A (en) * 1982-12-27 1985-05-07 International Business Machines Corporation Error recovery procedure using selective erasure
US4549234A (en) * 1981-02-04 1985-10-22 Sony Corporation Method and apparatus for tracking control
US4581567A (en) * 1982-05-28 1986-04-08 Citizen Watch Co., Ltd. Device for feeding magnetic head
US4604687A (en) * 1983-08-11 1986-08-05 Lucasfilm Ltd. Method and system for storing and retrieving multiple channel sampled data
US4620244A (en) * 1985-11-20 1986-10-28 Seagate Technology Compensation method to correct thermally induced off-track errors in disc drives
US4669004A (en) * 1986-02-27 1987-05-26 Quantum Corporation High capacity disk file with embedded sector servo
US4722085A (en) * 1986-02-03 1988-01-26 Unisys Corp. High capacity disk storage system having unusually high fault tolerance level and bandpass
US4726028A (en) * 1985-03-13 1988-02-16 Sony Corporation Method for error detection and error correction
US4730321A (en) * 1986-05-30 1988-03-08 Quantum Corporation Disk drive with improved error correction code
US4783705A (en) * 1986-02-27 1988-11-08 Quantum Corporation High capacity disk file with embedded sector servo and SCSI interface
US4817035A (en) * 1984-03-16 1989-03-28 Cii Honeywell Bull Method of recording in a disk memory and disk memory system
US4823212A (en) * 1986-11-26 1989-04-18 Hewlett-Packard Company Sampled servo code format and system for a disc drive
US4825403A (en) * 1983-05-16 1989-04-25 Data General Corporation Apparatus guaranteeing that a controller in a disk drive system receives at least some data from an invalid track sector
US4831609A (en) * 1986-10-21 1989-05-16 Csk Corporation Tracking method and apparatus for optical recording medium
US4914656A (en) * 1988-06-28 1990-04-03 Storage Technology Corporation Disk drive memory
US4942579A (en) * 1987-06-02 1990-07-17 Cab-Tek, Inc. High-speed, high-capacity, fault-tolerant error-correcting storage system

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE2622696C3 (de) * 1976-05-21 1979-09-13 Licentia Patent-Verwaltungs-Gmbh, 6000 Frankfurt Schaltungsanordnung zur Fehlerprüfung und Fehlerkorrektur von Binärdaten
IT1182420B (it) * 1985-02-08 1987-10-05 Olivetti & Co Spa Apparecchiatura per registrare e leggere informazioni su un disco magnetico

Patent Citations (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
USRE31166E (en) * 1967-08-09 1983-03-01 Rca Corporation Recovery of normally illegible, magnetically recorded information
US4103234A (en) * 1967-11-24 1978-07-25 General Dynamics Corp. System for transmission storage and/or multiplexing of information
US3943566A (en) * 1974-07-17 1976-03-09 International Business Machines Corporation Dynamic skew correction for rotating head magnetic recorder
US4024498A (en) * 1975-08-04 1977-05-17 Mcintosh Billy L Apparatus for dead track recovery
US4148082A (en) * 1976-09-13 1979-04-03 Sony Corporation Tracking control apparatus
US4092732A (en) * 1977-05-31 1978-05-30 International Business Machines Corporation System for recovering data stored in failed memory unit
US4445153A (en) * 1980-03-07 1984-04-24 Ricoh Company, Ltd. Method of correcting track misalignment of floppy disc and apparatus therefor
US4445145A (en) * 1980-11-07 1984-04-24 Sony Corporation Video signal reproducing apparatus
US4549234A (en) * 1981-02-04 1985-10-22 Sony Corporation Method and apparatus for tracking control
US4414587A (en) * 1981-12-17 1983-11-08 Sri International Skew sensing for digital tape playback
US4477849A (en) * 1982-04-12 1984-10-16 Pioneer Magnetics, Inc. Magnetic disc drive head alignment system
US4581567A (en) * 1982-05-28 1986-04-08 Citizen Watch Co., Ltd. Device for feeding magnetic head
US4516165A (en) * 1982-12-27 1985-05-07 International Business Machines Corporation Error recovery procedure using selective erasure
US4825403A (en) * 1983-05-16 1989-04-25 Data General Corporation Apparatus guaranteeing that a controller in a disk drive system receives at least some data from an invalid track sector
US4604687A (en) * 1983-08-11 1986-08-05 Lucasfilm Ltd. Method and system for storing and retrieving multiple channel sampled data
US4817035A (en) * 1984-03-16 1989-03-28 Cii Honeywell Bull Method of recording in a disk memory and disk memory system
US4726028A (en) * 1985-03-13 1988-02-16 Sony Corporation Method for error detection and error correction
US4620244A (en) * 1985-11-20 1986-10-28 Seagate Technology Compensation method to correct thermally induced off-track errors in disc drives
US4722085A (en) * 1986-02-03 1988-01-26 Unisys Corp. High capacity disk storage system having unusually high fault tolerance level and bandpass
US4669004A (en) * 1986-02-27 1987-05-26 Quantum Corporation High capacity disk file with embedded sector servo
US4783705A (en) * 1986-02-27 1988-11-08 Quantum Corporation High capacity disk file with embedded sector servo and SCSI interface
US4730321A (en) * 1986-05-30 1988-03-08 Quantum Corporation Disk drive with improved error correction code
US4831609A (en) * 1986-10-21 1989-05-16 Csk Corporation Tracking method and apparatus for optical recording medium
US4823212A (en) * 1986-11-26 1989-04-18 Hewlett-Packard Company Sampled servo code format and system for a disc drive
US4942579A (en) * 1987-06-02 1990-07-17 Cab-Tek, Inc. High-speed, high-capacity, fault-tolerant error-correcting storage system
US4914656A (en) * 1988-06-28 1990-04-03 Storage Technology Corporation Disk drive memory

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5465244A (en) * 1991-03-28 1995-11-07 Kabushiki Kaisha Toshiba Disc reproducing apparatus
US5511051A (en) * 1992-08-31 1996-04-23 Olympus Optical Co., Ltd. Optical recording/reproducing apparatus capable of multi-track access for simultaneous multirack recording/reproduction
US5745639A (en) * 1995-01-04 1998-04-28 International Business Machines Corporation Apparatus for improving the continuous supply or display time of full-motion video and other data
US6314433B1 (en) * 1998-06-12 2001-11-06 Hewlett-Packard Company Frame-based heroic data recovery
US6345366B1 (en) * 1998-09-24 2002-02-05 International Business Machines Corporation Method and device for error recovery
US6381088B1 (en) * 1998-11-06 2002-04-30 Acorn Technologies, Inc. Apparatus for developing a dynamic servo signal from data in a magnetic disc drive and method
US6525897B2 (en) 1998-11-06 2003-02-25 Acorn Technologies, Inc. Apparatus for developing a dynamic servo signal from data in a magnetic disc drive and method
USRE40413E1 (en) * 1998-11-06 2008-07-01 Purchased Patent Management Llc Method and apparatus for developing a dynamic servo signal from data
US6633521B1 (en) * 1999-08-27 2003-10-14 Fujitsu Limited Storage apparatus and servo recovery method of it
US6412088B1 (en) * 1999-12-02 2002-06-25 Maxtor Corporation Method and apparatus for using block reread
US20100162068A1 (en) * 2008-12-19 2010-06-24 Kabushiki Kaisha Toshiba Memory device

Also Published As

Publication number Publication date
JPH0792896B2 (ja) 1995-10-09
KR950013379B1 (ko) 1995-11-02
KR920010432A (ko) 1992-06-26
EP0485795A3 (en) 1993-02-10
JPH04265504A (ja) 1992-09-21
CA2050161C (en) 1995-10-10
MY108075A (en) 1996-08-15
CA2050161A1 (en) 1992-05-15
CN1023518C (zh) 1994-01-12
CN1061485A (zh) 1992-05-27
EP0485795A2 (en) 1992-05-20

Similar Documents

Publication Publication Date Title
US4914529A (en) Data disk defect handling using relocation ID fields
US5163162A (en) System and method for data recovery in multiple head assembly storage devices
US5189566A (en) Method and apparatus for recovering data
US5163136A (en) System for assembling playback data frames using indexed frame buffer group according to logical frame numbers in valid subcode or frame header
US5613088A (en) Raid system including first and second read/write heads for each disk drive
US4516165A (en) Error recovery procedure using selective erasure
US4916701A (en) Method and system for correcting long bursts of consecutive errors
EP0494669A1 (en) Method and apparatus for fault tolerant transducing in a memory file
US3641534A (en) Intrarecord resynchronization in digital-recording systems
US7965462B2 (en) Storing partial data sets to magnetic tape
US20050201015A1 (en) Data transfer device and method
US5966263A (en) Method and apparatus to increase head/AE select robustness for a system using multiple head servo write
US4527273A (en) Magnetic disc device
US5341479A (en) Address mark triggered read/write head buffer
US3771125A (en) Error correcting system of a magnetic tape unit
US5485476A (en) Method and system for error tolerant synchronization character detection in a data storage system
US8266499B2 (en) CRC protection of data stored in XOR buffer
US20040030737A1 (en) Algebraic decoder and method for correcting an arbitrary mixture of burst and random errors
US5623505A (en) Apparatus for detecting presence of information data for use in recording medium playing apparatus
EP0456690B1 (en) Read/write head buffer
USRE28265E (en) Svc out
JPH0620403A (ja) 磁気ディスク装置
JPS6040104B2 (ja) 磁気記録装置における記録検査方式
US9786320B2 (en) Systems and methods for missed media sector alignment
JPS59178665A (ja) 磁気テープ装置の位置決め制御方法

Legal Events

Date Code Title Description
AS Assignment

Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, ARMON

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNORS:BERRY, ROBERT L.;CENTERWALL, BRANDT C.;LUNING, STEPHEN G.;AND OTHERS;REEL/FRAME:005574/0526;SIGNING DATES FROM 19910103 TO 19910108

FPAY Fee payment

Year of fee payment: 4

REMI Maintenance fee reminder mailed
LAPS Lapse for failure to pay maintenance fees
FP Lapsed due to failure to pay maintenance fee

Effective date: 20001110

STCH Information on status: patent discontinuation

Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362